We are characterizing a number of ego (enhancer of
glp-1) genes at the molecular level. Two genes,
ego-1 (Qiao et al. 1995) and
ego-6 (J. Spoerke, M. Klein, and E. Maine, unpublished), map between
goa-1 and
gld-1 (most likely to the right of
lrp-1) and show intergenic non-complementation (i.e.,
ego-6 +/ +
ego-1 animals are mutant). The only genetic evidence to suggest that
ego-6 and
ego-1 are different genes is deficiency mapping data; two independently derived chromosomal deficiencies, nDf25 and mnDf111, uncover
ego-1 alleles (
om18,
om71) but not
ego-6 alleles (
om54,
om58,
om84,
om96,
om97,
om119).
ego-6 mutations cause a variety of germline defects including enhancement of
glp-1(ts) and
lag-1(ts), premature onset of meiosis, slow progression through the early meiotic "transition zone", and various gametogenesis defects.
ego-1 mutations cause a similar, but not identical, phenotype.
ego-6 +/ +
ego-1 transheterozygotes resemble
ego-6 mutants. To determine whether
ego-6 and
ego-1 indeed are different genes and to better understand their role(s) in germline development, we have begun molecular studies. Using cosmids spanning the
goa-1 to
gld-1 region, we looked for DNA rearrangements associated with
ego-6 alleles. Using cosmid F26A3 as a probe, we detected a deletion of ~300 bp associated with the UV-induced
ego-6(
om84) allele. It maps to the vicinity of the predicted transcription unit, "F26A3.3". F26A3 was not able to detect polymorphisms in any other
ego-6 mutants and none of the other cosmids detected RFLPs in any of the
ego-6 mutants. Earlier studies using cosmids spanning the
lrp-1 to
gld-1 region failed to detect any polymorphisms associated with
ego-1 mutations (S. Stacey and E. Maine, unpublished data). We isolated and sequenced cDNAs spanning the F26A3.3 region and carried out RNA blots with several subfragments of F26A3 as probes. The GeneFinder prediction for this region was fairly accurate, but the large predicted transcript, F26A3.3, actually turned out to encode two independent genes. Interestingly, the two genes are structurally related and encode "novel" proteins that are ~58% identical at the amino acid level (as predicted from cDNA sequences). Based on our studies, the
om84 deletion interrupts the coding region of the upstream gene; we presume that this gene is
ego-6. C. elegans appears to contain at least one other relative based on GeneFinder predictions; a loosely related gene is predicted by the S. pombe genome project. In-progess experiments aim to (1) confirm the identity of
ego-6, (2) investigate whether
ego-6 or the downstream, related gene might be
ego-1, and (3) examine the tissue-specificity of
ego-6 expression. To examine question (1), we are amplifying and sequencing the putative
ego-6 gene from other
ego-6 mutant strains. We are also sequencing
om84 to determine whether the deletion shifts the open reading frame (if so,
om84 is likely a null allele). To investigate question (2), we are amplifying the putative
ego-6 gene and its downstream relative from
ego-1 mutant strains to determine if either gene contains mutations associated with the
ego-1 alleles. If
ego-6 and
ego-1 are identical, then the upstream gene should contain (a) mutation(s). Alternatively, if
ego-6 and
ego-1 are different genes, then it is possible that the duplicated gene is
ego-1; if so, then the downstream gene should contain (a) mutation(s). To investigate question (3), we are examining
glp-4(
bn2ts) mutants (raised at 25 C) for the presence of
ego-6 RNA. Based on its mutant phenotype, we suspect that
ego-6 might be expressed specifically in the germ line.
glp-4(
bn2ts) mutants have very small germ lines, and have been widely used to assay for germline-specific (or -enriched) expression of a variety of genes. Results of these three lines of investigation will be presented at the meeting.